I’m delighted to share a few new features with you! We’ve added workspaces, improved the run table, and made reports more organized.
You’ll notice a new workspace bar at the bottom of the project and run pages. Your workspace auto-saves your configuration of charts, filters, and table settings. It’s a sandbox for you to quickly compare runs and dig into your data. To save a snapshot of your workspace, click “Export to Report.” You'll be able to share the report with collaborators and write markdown notes to annotate your visualizations. Here's an example project using the iNaturalist dataset.
New people opening the project for the first time will get their own workspace. You can provide a preset for new users. If you set up a particularly useful layout, you can configure it to serve as the default for new workspaces for others.
The run table is now more flexible and customizable. You can bulk tag or delete runs; you can resize columns and reorder them via drag-and-drop. The resulting table state will be autosaved in your project workspace. If you'd like to play with this table you can visit Stacey's iNaturalist example project.
In the table, we’ve cleaned up the UI for filtering, grouping, and sorting, and we’re introducing a brand new feature. Click the Magic Columns button to banish blank and repetitive columns, narrowing the view to the most relevant parameters and outcomes of your experiments.
Dozens of columns in a table can get unwieldy. Use our column editor to manage which columns you’d like to show.
We’ve improved the visual style of reports. Now it’s easier than ever to annotate your results and share them with colleagues. Check out Stacey’s iNaturalist report to see how she tried different approaches and visualized her results.
Enter your email to get updates about new features and blog posts.
We're building lightweight, flexible experiment tracking tools for deep learning. Add a couple of lines to your python script, and we'll keep track of your hyperparameters and output metrics, making it easy to compare runs and see the whole history of your progress. Think of us like GitHub for deep learning.
We are building our library of deep learning articles, and we're delighted to feature the work of community members. Contact Carey to learn about opportunities to share your research and insights.